ORIGINAL INVESTIGATION

Reinforcing and neurochemical effects of the “bath salts”constituents 3,4-methylenedioxypyrovalerone (MDPV)and 3,4-methylenedioxy-N-methylcathinone (methylone)in male rats

Charles W. Schindler1 & Eric B. Thorndike1 & Steven R. Goldberg1 & Kurt R. Lehner2 &

Nicholas V. Cozzi3 & Simon D. Brandt4 & Michael H. Baumann2

Received: 29 May 2015 /Accepted: 29 July 2015# Springer-Verlag Berlin Heidelberg (outside the USA) 2015

AbstractRationale 3,4-Methylenedioxypyrovalerone (MDPV) and 3,4-methylenedioxy-N-methylcathinone (methylone) are syn-thetic drugs found in so-called Bbath salts^ products. Bothdrugs exert their effects by interacting with monoamine trans-porter proteins. MDPV is a potent uptake blocker at trans-porters for dopamine and norepinephrine while methylone isa non-selective releaser at transporters for dopamine, norepi-nephrine, and serotonin (5-HT).Objectives We hypothesized that prominent 5-HT-releasingactions of methylone would render this drug less reinforcingthan MDPV.Methods To test this hypothesis, we compared behavioral ef-fects of MDPV and methylone using intravenous (i.v.) self-administration on a fixed-ratio 1 schedule in male rats. Addi-tionally, neurochemical effects of the drugs were examinedusing in vivo microdialysis in nucleus accumbens, in a sepa-rate cohort of rats.

Results MDPV self-administration (0.03 mg/kg/inj) was ac-quired rapidly and reached 40 infusions per session, similar tothe effects of cocaine (0.5mg/kg/inj), by the end of training. Incontrast, methylone self-administration (0.3 and 0.5 mg/kg/inj) was acquired slowly, and response rates only reached 20infusions per session by the end of training. In dose substitu-tion studies, MDPVand cocaine displayed typical inverted U-shaped dose-effect functions, but methylone did not. In vivomicrodialysis revealed that i.v. MDPV (0.1 and 0.3 mg/kg)increased extracellular dopamine while i.v. methylone (1 and3 mg/kg) increased extracellular dopamine and 5-HT.Conclusions Our findings support the hypothesis that eleva-tions in extracellular 5-HT in the brain can dampen positivereinforcing effects of cathinone-type drugs. Nevertheless,MDPVand methylone are both self-administered by rats, sug-gesting these drugs possess significant abuse liability inhumans.

Keywords 3,4-Methylenedioxypyrovalerone (MDPV) .

Methylone . Cocaine . Self-administration .Microdialysis .

Rats

Introduction

The non-medical use of psychoactive synthetic cathinones,commonly known as Bbath salts,^ has increased worldwidein recent years (Baumann 2014; German et al. 2014), oftenleading to life-threatening medical consequences (Kesha et al.2013; Wright et al. 2013). Three of the most popular syntheticcathinones, 3,4-methylenedioxypyrovalerone (MDPV), 4-methyl-N-methylcathinone (mephedrone), and 3,4-methylenedioxy-N-methylcathinone (methylone), have beenclassified by the USA as Schedule I controlled substances

* Charles W. Schindlercschind@helix.nih.gov

1 Preclinical Pharmacology Section, Intramural Research Program ofthe National Institute on Drug Abuse, National Institutes of Health,Baltimore, MD 21224, USA

2 Designer Drug Research Unit, Intramural Research Program of theNational Institute on Drug Abuse, National Institutes of Health,Baltimore, MD 21224, USA

3 Department of Cell and Regenerative Biology, University ofWisconsin School of Medicine and Public Health,Madison, WI 53706, USA

4 School of Pharmacy and Biomolecular Sciences, Liverpool JohnMoores University, Byrom Street, Liverpool, L3 3AF, UK

PsychopharmacologyDOI 10.1007/s00213-015-4057-0

(Drug Enforcement Administration 2011, 2013). The 36thmeeting of the Expert Committee on Drug Dependence, con-vened by the World Health Organization, was followed by arecommendation to place these three substances in Schedule IIof the UnitedNations Convention on Psychotropic Substances1971 (World Health Organization 2015). Despite these intro-ductions of legislative control, MDPVand methylone remainavailable in the street drug marketplace and are continuing tobe abused in the USA and elsewhere (Drug Enforcement Ad-ministration 2014; Seely et al. 2013). As a result, it is impor-tant to understand the pharmacological actions of these twodrugs, particularly with regard to their abuse liability.

MDPV and methylone are both classified as syntheticcathinones, but the drugs are chemically distinct and theirmolecular mechanisms of action differ. With respect to chem-ical structure, MDPV possesses a nitrogen-containing pyrrol-idine ring and an extended α-carbon alkyl chain, whereasmethylone does not. MDPV is an uptake blocker at trans-porters for dopamine (DAT) and norepinephrine (NET)(Baumann et al. 2013; Cameron et al. 2013; Eshleman et al.2013; Simmler et al. 2013), similar to cocaine but much morepotent. Importantly, MDPV is highly selective for DAT andNET, with little affinity for the serotonin transporter (SERT).In contrast, methylone acts as a non-selective substrate forDAT, NET, and SERT, thereby causing the transporter-mediated release of all three monoamine transmitters, similarto the actions of 3,4-methylenedioxymethamphetamine(MDMA) (Baumann et al. 2012; Eshleman et al. 2013;Simmler et al. 2013). MDPVand methylone increase locomo-tor activity in rodents (Aarde et al. 2013; Gatch et al. 2013;Lopez-Arnau et al. 2012; Marusich et al. 2012) and can in-crease body temperature, particularly after high-dose admin-istration or at elevated ambient temperatures (Fantegrossi et al.2013; Lopez-Arnau et al. 2014a, b; Kiyatkin et al. 2015). Bothdrugs also enhance intracranial self-stimulation (Bonano et al.2014; Watterson et al. 2012, 2014) and engender conditionedplace preference (Karlsson et al. 2014), effects that are indica-tors of high abuse potential. In studies that have directly com-pared the in vivo effects of MDPVand methylone, MDPV istypically 3–10 times more potent than methylone (Bonanoet al. 2014; Gatch et al. 2013; Karlsson et al. 2014; Marusichet al. 2012; Kiyatkin et al. 2015).

The pharmacological effects of methylone might be ex-pected to mimic those produced by MDMA since the twodrugs share similar chemical structures and the ability to re-lease 5-HT from neurons. MDMA is self-administered byanimals, but its reinforcing effects are weak when comparedto other psychostimulants, most likely due to its stimulation of5-HT release (De La Garza et al. 2007; Schenk et al. 2007;Bradbury et al. 2014). Accumulating preclinical evidenceshows that elevations in extracellular 5-HT in the brain candampen locomotor and reinforcing effects mediated by dopa-minergic stimulants (Wee and Woolverton 2006; Baumann

et al. 2011; Bauer et al. 2013). The fact that MDPV has rela-tively specific effects on dopamine and norepinephrine, with-out affecting 5-HT systems, would lead one to predict greaterself-administration of this drug when compared to methylone.Somewhat surprisingly, methylone (Watterson et al. 2012;Creehan et al. 2015) and MDPV (Aarde et al. 2013, 2015;Watterson et al. 2014) are both readily self-administered byrats, but methylone appears to be less potent and supportssomewhat lower rates of responding than MDPV.Importantly, Creehan et al. (2015) reported much lower ratesof responding for methylone in female rats than Wattersonet al. (2012) reported in male rats, suggesting that there maybe sex differences in responsiveness to the drug.

Methodological differences across the various studies ex-amining self-administration of MDPV and methylone makedirect comparison of the results difficult. For example, in stud-ies investigating the reinforcing effects of MDPV in male rats(Aarde et al. 2013; Watterson et al. 2014), subjects were firstpre-trained with food prior to drug self-administration, whilethis was not the case for methylone (Watterson et al. 2012). Inaddition, Creehan et al. (2015) used female rats to studymethylone self-administration. Here, we wished to carry outa side-by-side comparison of MDPV and methylone self-administration in male rats, without pre-training for food,and under identical conditions to allow for direct comparisonof reinforcing effects. In addition, we used in vivo microdial-ysis in male rats to compare the effects of the two drugs onextracellular concentrations of dopamine and 5-HT in the nu-cleus accumbens. Based on our previous neurochemical find-ings (Baumann et al. 2012; 2013), we predicted that MDPVadministration would increase extracellular concentrations ofdopamine only, while methylone would increase concentra-tions of both dopamine and 5-HT. Given the stronger seroto-nergic effects of methylone, we hypothesized that self-administration of this drug would be less robust when com-pared to MDPV.

Materials and methods

Subjects Male Sprague-Dawley rats, weighing 300–400 g atthe beginning of the studies, were used for these experiments.They were housed under standard vivarium conditions of con-trolled temperature (22±2°C) and humidity (45±5 %) withfood and water freely available. Rats were group-housed,two or three per cage, prior to surgery but single-housed there-after. Rats for self-administration studies were housed under a12:12 h reverse light-dark cycle (lights on at 2200 hours),whereas rats for microdialysis studies were housed under a12:12 h normal light-dark cycle (lights on at 0700 hours).Guidelines of the Institutional Animal Care and Use Commit-tee at the National Institute onDrugAbuse (NIDA)/Intramural

Psychopharmacology

Research Program (IRP) and the Guide for the Care and Useof Laboratory Animals were followed at all times.

Drugs 3,4-Methylenedioxypyrovalerone (MDPV) and 3,4-methylenedioxy-N-methylcathinone (methylone) were syn-thesized in racemic form as HCl salts in our laboratories.Chemical and structural analysis included proton nuclearmagnetic resonance, gas chromatography/mass spectrometry,thin layer chromatography, and melting point determination.All data confirmed the expected structures. Cocaine hydro-chloride (NIDA/IRP, Baltimore, MD), MDPV, and methylonewere all dissolved in sterile saline prior to administration.

Self-administration studies Rats were anesthetized by acombination of 100 mg/kg ketamine and 10 mg/kg xylazine,(i.p.), and intravenous (i.v.) jugular catheters were implantedaccording to procedures described previously (Schindler et al.2011). Briefly, 3 cm of Silastic tubing (Dow Corning,0.44 mm i.d., 0.9 mm o.d.) was inserted into the right jugularvein and connected to vinyl tubing (Dural Plastics, 0.5 mmi.d., 1.0 mm o.d.) that exited the back at the midscapular re-gion and was plugged with an obturator. Immediately follow-ing catheter implantation, an incision was made posterior tothe exit site of the catheter and a 20-mm nylon screw wasinserted serving as a back mount. A metal spring, whichhoused Tygon extension tubing connecting the fluid swivel(Instech, Plymouth Meeting, PA) on the top of the cage tothe i.v. catheter, was attached to this mount at the time ofself-administration sessions. This tubing was attached to a10-ml syringe controlled by a motor-driven syringe pump out-side the sound attenuation cubicle. The pump (PHM-100,Med Associates) delivered a constant volume (0.2 ml/inj) ofdrug solution over 2 s. Concentration of drug was adjusted foreach rat’s bodyweight to give the desired dose. Catheters wereflushed before and after each training session with 0.1 ml of asaline solution containing 1.25 units/ml heparin and0.08 mg/ml gentamicin. If a catheter failed during the study,a second catheter was implanted on the opposite side. Afterthe second catheter failure, the rat was removed from thestudy.

Ten training chambers were used (ENV-008CT, Med As-sociates, St. Albans VT). Each chamber was enclosed in asound-attenuation cubicle equipped with a fan to provide ven-tilation and stable background noise (ENV-018M, Med Asso-ciates). Each chamber had a grid floor and two nose-pokeresponse holes (ENV-114BM, Med Associates), one on eachside of a food trough (not used in this study). The nose-pokeholes could be illuminated from inside the hole by a dimyellow light. A houselight (ENV-215M, Med Associates)was situated above the nose-poke holes. Experimental eventswere controlled by a MED-PC computer system (MedAssociates).

Following surgery and at least 7 days of recovery, the cath-eters were connected to the extension tubing, and the rats wereplaced in the training chambers every weekday. Sessions be-gan with the illumination of the houselight and the two nose-poke holes. A nose-poke in the active hole was reinforced on afixed-ratio 1 schedule with an infusion of drug, and the house-light and nose-poke hole lights were turned off and a 20-stimeout began. Nose-pokes in the inactive hole were recorded,but had no scheduled consequence. During the timeout, re-sponses were recorded, but not reinforced. Following thetimeout, the houselight and nose-poke hole lights were turnedon and active nose-pokes were again reinforced. Whether theright or left nose-poke was active was counterbalanced acrosssubjects. Sessions lasted for 2 h. Initial acquisition doses were0.03mg/kg/inj for MDPV, 0.3 or 0.5 mg/kg/inj for methylone,and 0.5 mg/kg/inj for cocaine. The training doses of MDPVand methylone were chosen based on the findings of others(e.g., Watterson et al. 2012, 2014) while the training dose ofcocaine was based on our previous experience. Most rats re-ceived at least 10 days of acquisition training, except those ratstrained on 0.5 mg/kg/inj methylone where most rats received17 days of training.

Following acquisition training, dose-effect testing beganfor all the groups except those originally trained on0.5 mg/kg/inj methylone. The dose of each drug was variedto three additional doses. Each dose was available for self-administration for 3 days, with results for days 2 and 3 aver-aged for analysis. Following dose-effect testing, the rats werereturned to the original training dose. After responding stabi-lized, saline was substituted for each drug and training contin-ued until responding decreased.

Microdialysis studies Rats were anesthetized with sodiumpentobarbital (60 mg/kg, i.p.), and each rat received a surgi-cally implanted jugular catheter and an intracranial guide can-nula as previously described (Baumann et al. 2012). Briefly,approximately, 3 cm of Silastic tubing was inserted into theright jugular vein and connected to vinyl tubing that exited theback at the midscapular region and was plugged with an ob-turator. Immediately following catheter implantation, the ratwas placed into a stereotaxic apparatus and an intracerebralguide cannula (CMA/12, Harvard Apparatus, Holliston, MA)was surgically implanted. The guide cannula was aimed at thenucleus accumbens, according the coordinates: 1.7 mm lateraland 1.6 mm anterior to bregma (reference zero) and 6.0 mmventral to the dura surface. Rats were allowed at least 1 weekfor recovery from surgery. On the evening before an experi-ment, a 2×0.5 mm dialysis probe (CMA/12, Harvard Appa-ratus) was inserted into the guide cannula, and an extensiontube was attached to the jugular catheter. Each rat was placedinto its own enclosure and connected to a tethering system.The enclosure was equipped with photobeams to detect loco-motor activity (Truscan, Harvard Apparatus). Probes were

Psychopharmacology

perfused overnight with Ringer’s salt solution pumped at aflow rate of 0.6 μl/min. On the next morning, dialysate sam-ples were collected at 20-min intervals. Samples were imme-diately assayed for dopamine and 5-HT by HPLC-ECD asdescribed elsewhere (Baumann et al. 2011).

Rats were randomly assigned to groups receiving eitherdrug or saline injections. Once three stable baseline sampleswere obtained, rats received two sequential i.v. injections ofdrug: one dose at time zero, followed by a 3-fold higher dose60 min later. Drugs were administered i.v. to mimic the self-administration route, and the low drug doses were ~3-foldgreater than the training dose self-administered. For MDPV,rats received 0.1 mg/kg followed by 0.3 mg/kg. Formethylone, rats received 1.0 mg/kg followed by 3 mg/kg.Control rats received sequential i.v. injections of saline(1 ml/kg) according to the same schedule. Microdialysis sam-ples were collected every 20min throughout the post-injectionperiod for 120 min. At the end of the experiments, rats wereeuthanized with CO2 and decapitated. Brains were cut on acryostat (CM1950, Leica Biosystems, Buffalo Grove, IL) andexamined to verify placement of microdialysis probe tipswithin the nucleus accumbens. Only those rats with correctplacements were included in data analyses.

Data analysis For self-administration studies, only responsesduring the reinforcement period were used for analysis.Timeout responses were not included, but were generally verylow following the first couple of days of training. Data fromthe self-administration experiments were analyzed using a lin-ear mixed-effects model. For the acquisition and extinctionanalysis, day and response (active/inactive) were treated aswithin-subject factors. For the dose-effect study, dose wastreated as a within-subject factor. Dose/h was calculated bymultiplying the total number of infusions by the dose/infusionand dividing by the 2 h session time. For microdialysis stud-ies, all neurotransmitter concentrations (pg/μl) and motor ac-tivity measures (cm/20 min) were normalized to a percentageof control values based on three pre-injection baseline samplesand the corresponding 20-min activity bins. Microdialysis da-ta were analyzed using two-way analysis-of-variance (treat-ment×time). Post hoc analyses were performed using t testswith Bonferroni corrections.

Results

Self-administration Figure 1 shows acquisition data forMDPV (0.03 mg/kg/inj), methylone (0.3 mg/kg/inj), and co-caine (0.5 mg/kg/inj). Both active/reinforced and inactive re-sponses are shown. Timeout responses are not included. Onthe first day of training, rats took over 40 injections of MDPV(43.2±10.2) and maintained that level of responding through-out training. On the first day of training, inactive responses

were high as well (29.4±9.5), but decreased rapidly after day1. Statistical analysis showed a significant effect of the inter-action of response (active/inactive) and day (F9,312=3.17,p<.01). Post hoc analysis showed that active and inactiveresponses were significantly different (solid symbols inFig. 1) for every day beyond day 1 (p<.001).

In contrast to MDPV, reinforced responses for methylonewere below 20 on day 1 (14.4±3.3) and slowly increased overtime to a maximum of 25.5±4.9 by day 10. Inactive responseswere similar to active responses over the first few days oftraining (8.5±1.8 on day 1), but a separation between activeand inactive responses did eventually develop. Statistical

Fig. 1 Acquisition and extinction of self-administration of 0.03 mg/kg/inj MDPV (top panel, acquisition n=17–20, extinction n=4–12),0.3 mg/kg/inj methylone (middle panel, acquisition n=15, extinctionn=6–9), and 0.5 mg/kg/inj cocaine (bottom panel, acquisition n=14–16, extinction n=7–9). The first 10 days of acquisition are shown, follow-ed by the first 6 days of extinction. Active responses and inactive re-sponses do not include timeout responses. Therefore, active responsesare equivalent to infusions. Filled symbols for active responses are sig-nificantly different from inactive on the same day. A acquisition, Eextinction

Psychopharmacology

analysis again showed a significant interaction of response(active/inactive) and day (F9,246=3.14, p<.001). Post hocanalysis showed that active and inactive response differed ondays 7–10 (solid symbols, p<0.01).

For cocaine, active/reinforced responses were initially sim-ilar to those of MDPV, averaging 32.1±11.6 active responseson day 1. Inactive responses were also high initially (38.0±5.7on day 2), but dropped over the ensuing days of training.Because of the variability in the inactive response data, thestatistical analysis failed to show any significant effects.Whenonly active responses were compared across the 3 drugs, sta-tistical analysis showed a significant effect of drug (F2,47=7.8,p<0.003), but not for either the day factor or the interaction.Post hoc analysis showed all three drugs differed (cocaine vsMDPV, p<0.01, cocaine vs methylone and MDPV vsmethylone, p<0.001). Active responses for MDPV wereslightly but consistently higher for MDPV when comparedto cocaine. Active responses for methylone were consistentlywell below those for both MDPVand cocaine.

Watterson et al. (2012) reported much higher levels of self-administration for methylone than were observed here. At adose of 0.5 mg/kg/inj, they reported rates of up to 80 infusionsper 2 h session by day 10 of training, and this level ofresponding continued throughout the 20 days of training. Tomore closely replicate the procedure of Watterson et al. (2012),we trained a separate group of rats on 0.5 mg/kg/inj methyloneand continued training beyond 10 days. Due to catheter limita-tions, most rats were only able to continue training for 14–17 days. Figure 2 shows the results for the 0.5 mg/kg/inj group.Over the first 10 days of training, the results for 0.5 mg/kg/injare comparable to those for the 0.3 mg/kg/inj methylone dose(see Fig. 1). After 10 days, there was only a small increase inthe number of injections for the 0.5 mg/kg/inj group. Statisticalanalysis showed a significant interaction of response (active/inactive) and day (F16,244=6.3, p<0.0001). Post hoc analysisshowed active and inactive responses differed on days 6, 7, and9–17 (solid symbols in Fig. 2).

When saline was substituted for drug, responding decreasedfor MDPV, methylone, and cocaine (Fig. 1). Respondingdropped rapidly for MDPV, but relatively slowly for cocaine.Formethylone, there was an immediate decrease in responding.Statistical analysis showed that for MDPV the interaction ofresponse (active/inactive) and day was significant (F5,80=3.8,p<.01) with post hoc tests showing that only the difference inactive/inactive responses on day 1 was significant (solid sym-bol). For methylone, neither the main effects of response andday nor the interaction was significant. For cocaine, the maineffects were both significant (day F5,76=4.47, p<.01; responseF1,8=18.14, p<.01), but the interaction was not.

Figure 3 shows the results for the dose-effect manipulation.The top panel shows the average number of reinforced re-sponses. For comparison, the average number of saline infu-sions for the last day of extinction is shown for each drug. For

both MDPV and cocaine, results typical for stimulant drugself-administration were observed, with peak infusions occur-ring at lower doses and rates of responding decreasing withincreases in dose. However, for methylone, responding wasrelatively flat across doses. For all three drugs, there was asignificant effect of dose (MDPV F4,20=21.95, p<.0001;methylone F4,27=3.79, p<0.05; cocaine F4,32 =22.9,

Fig. 2 Acquisition of self-administration of a higher dose of methylone0.5 mg/kg/inj (n=5–9). Training for the higher dose of methylone wascontinued beyond 10 days and the first 17 days of acquisition is shown.Filled symbols for active responses are significantly different frominactive responses on the same day. For comparison, the results for0.3mg/kg/inj methylone fromFig. 1 are shown as solid (active responses)and dashed (inactive responses) lines

Fig. 3 Dose-effect functions for infusions (top panel) and intake in dose/h (bottom panel) is shown for MDPV (n=4–11), methylone (n=4–13),and cocaine (n=7–12). Dose was manipulated following initial acquisi-tion of the self-administration response. Solid symbols in the top panelindicate differences from the appropriate saline (S) point. Solid symbols inbottom panel are significantly different from open symbols for same drug,but do not differ from each other for the same drug

Psychopharmacology

p<0.0001). The number of infusions taken was significantlyabove saline for multiple doses for both MDPV (0.003, 0.01,and 0.03 mg/kg/inj, solid symbols) and cocaine (0.1,0.3 mg/kg/inj, solid symbols), but for methylone, only the0.1 mg/kg/inj dose maintained infusion rates above saline(p<0.05, solid symbol).

The bottom panel of Fig. 3 shows the average hourly intakefor each drug. For cocaine, the pattern of responding led tofairly steady rates of intake across doses, while for methylonerates of intake increased with increasing dose. Rate of intakefor MDPV increased most dramatically from 0.003 mg/kg/inj,a dose that did not reliably maintain responding, to0.01 mg/kg/inj. Above 0.01 mg/kg/inj, intake increased grad-ually. Again, for all three drugs, there was a significant effectof dose (MDPV F3,14=15.37, p<0.0001; methylone F3,19=35.72, p<0.0001; cocaine F3,24=14.4, p<0.0001). For bothMDPV and cocaine, the intake at the lowest dose was signif-icantly different from the other doses (p<0.01, open symbol),but the higher doses did not differ from each other. Formethylone, the 0.01 and 0.03 doses did not differ from eachother and the 0.1 and 0.3 doses did not differ. All other com-parisons were significant (p<0.01).

Microdialysis Figure 4 depicts the effects of MDPVor salineinjections on extracellular concentrations of dopamine and 5-HT in nucleus accumbens of conscious rats, along with corre-sponding locomotor activity. MDPV significantly increased di-alysate dopamine at both doses tested (F8,99=11.68,p<0.0001), but had no effect on dialysate 5-HT (F8,99=0.2).The elevations in dopamine produced by MDPV were 2-foldand 4-fold above basal preinjection values after the 0.1 and0.3 mg/kg doses, respectively. As shown in the bottom panelof Fig. 4, MDPV markedly increased locomotion in a dose-dependent manner (F8,99=21.17, p<0.0001), and motor activa-tion was temporally correlated with elevations in extracellulardopamine. Figure 5 shows the effects of methylone injectionson extracellular concentrations of dopamine and 5-HT. In con-trast to MDPV, methylone significantly increased concentra-tions of both dopamine (F8,99=5.32, p<0.0001) and 5-HT(F8,99=11.92, p<0.0001), with the relative increases in 5-HTbeing much greater. For example, the rise in dialysate 5-HTafter 3.0 mg/kg was 8-fold above basal preinjection values,while the corresponding elevation in dialysate dopamine wasonly about 3-fold. Methylone produced a dose-related increasein locomotor activity (F8,99=12.05, p<0.01), though this effectwas smaller in magnitude when compared to MDPV.

Discussion

A major aim of the present study was to directly compare thereinforcing and neurochemical effects of the bath salts constit-uents MDPV and methylone. Although these drugs are both

classified as synthetic cathinones, they are quite different interms of their selectivity and mechanism at monoamine trans-porters. MDPV is a potent uptake blocker at DAT and NETwith minimal activity at SERT (Baumann et al. 2013; Camer-on et al. 2013; Eshleman et al. 2013; Simmler et al. 2013),whereas methylone is a non-selective transporter substrate thatevokes the release of dopamine, norepinephrine and 5-HT(Baumann et al. 2012; Eshleman et al. 2013; Simmler et al.2013). We hypothesized that the serotonergic effects ofmethylone would render this drug less reinforcing when com-pared to MDPV. It was found that rats acquired MDPV self-administration rapidly, within the first few days of training,and the drug maintained high rates of responding (40

Fig. 4 Effects of MDPV (n=7) or saline vehicle (N=7) on extracellularconcentrations of dopamine (DA, top panel) and serotonin (5-HT, middlepanel) in the nucleus accumbens, and on horizontal locomotor activity(activity, bottom panel), in conscious rats undergoing microdialysis.Drugs or saline were administered i.v. at the indicated times (arrows)and doses (mg/kg). *p<0.05, **p<0.01 = significant difference fromsaline (n=6) at the corresponding time point

Psychopharmacology

infusions per 2-h session). The acquisition of methylone, onthe other hand, occurred slowly over the course of many days,and the number of responses (20 infusions per 2-h session)was much less than for MDPV. In dose-substitution studies,MDPV and cocaine gave typical inverted U-shaped dose-ef-fect functions, but methylone did not. Microdialysis findingsrevealed that MDPV produces selective increases in dialysatedopamine while methylone increases dialysate dopamine and5-HT, with larger effects on 5-HT. Collectively, the presentfindings support the hypothesis that elevations in central 5-HT can dampen reinforcing effects mediated by mesolimbicdopamine.

In the present study, we noted robust self-administration ofMDPV from the very beginning of training. High rates of drugintake were observed with clear discrimination between activeand inactive responses by day 2 of training. The MDPV dose-effect function was typical of psychomotor stimulants, withresponse rates peaking at 0.01 mg/kg/inj. At higher doses,response rates decreased while intake increased only slightly.The results shown here confirm previous reports of avid self-administration ofMDPVinmale rats (Aarde et al. 2013, 2015;Watterson et al. 2014). In terms of absolute response rates, wefound that MDPVmaintained rates comparable to cocaine (40infusions per 2 h session), although MDPV was about 10-times more potent than cocaine. The high rates of infusionsfor MDPV observed here are similar to those reported byAarde et al. (2013), but less than those reported by Wattersonet al. (2014) who found rates around 60 infusions per 2 hsession for an MDPV training dose of 0.05 mg/kg/inj. Wedemonstrate here that rats tend to regulate their intake ofMDPV and cocaine, with intake for both drugs not differingsignificantly at the three highest doses tested. The presentinverted U-shaped dose-effect functions with MDPV and co-caine are consistent with our previous results using cocaine asa training drug in rats (Panlilio et al. 2003). When saline wassubstituted for drug, rates of responding for both MDPV andcocaine decreased, and the discrimination between the activeand inactive responses was lost. One potential caveat in ourdata set is that cocaine-trained rats exhibited many more inac-tive responses when compared to MDPV-trained rats. A closeinspection of the results showed that variability for inactiveresponses in cocaine-trained rats was due primarily to certainrats that showed high levels of inactive responses on individ-ual days of training, particularly day 2. We observed that in-active responses were rare among the rats responding forMDPVafter the first day of training.

In contrast to MDPV and cocaine, we found thatresponding for methylone was low initially, with little discrim-ination between active and inactive responses. Respondingdid slowly increase over the course of several days, and dis-crimination between active and inactive responses eventuallyemerged. However, even when the dose of methylone wasincreased from 0.3 to 0.5 mg/kg/inj, and additional days oftraining were allowed, responding remained low relative toMDPV and cocaine. When saline was substituted formethylone, responding immediately decreased and the dis-crimination between the responses was lost. In general, lessresponding was seen for methylone in the present experiments(i.e., 20 infusions per 2 h session) than in a previous studyusing methylone at comparable training doses (i.e., up to 80infusions per 2 h session) (Watterson et al. 2012). The reasonfor this discrepancy is not clear since both studies used thesame male rat strain, trained rats during the dark phase of thelight cycle (i.e., active phase), omitted food pre-training, andhad similar schedule parameters. Watterson et al. (2012) used

Fig. 5 Effects of methylone (n=7) or saline vehicle (N=7) onextracellular concentrations of dopamine (DA, top panel) and serotonin(5-HT,middle panel) in the nucleus accumbens, and horizontal locomotoractivity (activity, bottom panel), in conscious rats undergoingmicrodialysis. Drugs or saline were administered i.v. at the indicatedtimes (arrows) and doses (mg/kg). *p<0.05, **p<0.01 = significantdifference from saline (n=6) at the corresponding time point

Psychopharmacology

lever press responses in their study, rather than the nose-pokeresponses used in our study, but nose-poke responding typi-cally facilitates, rather than suppresses, training. Wattersonet al. did find that methylone was not as effective in maintain-ing responding as MDPV when compared across studies(Watterson et al. 2012, 2014). Furthermore, it should be notedthat a recent study by Creehan et al. (2015) demonstrated thatfemale rats acquire methylone self-administration in a slowand steady manner, over the course of 2 weeks of training,with response rates in the range of 20 infusions per 2 h session.Thus, the recent methylone data from female rats reported byCreehan et al. agree closely with what we found here in malerats. In our study, when the methylone dose was manipulated,little change in responding was observed, leading to largechanges in total intake.Watterson et al. (2012) trained separategroups of rats at different doses of methylone rather than ma-nipulating dose after initial training. They found clearer dif-ferences in responding between doses than observed here,even over a similar dose range. Nevertheless, whileresponding for methylone was low in our study, self-administration behavior was observed, just not to the sameextent as that of MDPVand cocaine. With regard to potency,the present data agree with published self-administration stud-ies in rats, which consistently find that methylone is about 10-times less potent than MDPV (Aarde et al. 2013, 2015; Wat-terson et al. 2012, 2014).

The microdialysis data reported here may provide clues tothe neurochemical mechanisms underlying differences in thebehavioral effects of MDPV versus methylone. However, itshould be noted that microdialysis experiments were carriedout during the light phase of the circadian cycle whereas self-administration experiments were carried out during the darkphase. The different timing for the neurochemical and behav-ioral studies represents a limitation to direct comparison of theresults. We found that MDPV and methylone both producerapid elevations in extracellular dopamine upon i.v. adminis-tration. In fact, the rise in dialysate dopamine produced by0.1 mg/kg MDPV was nearly identical to the rise producedby 1.0 mg/kg methylone (i.e., 2-fold increase). On the otherhand, MDPV had no discernable effect on extracellular 5-HTwhile methylone produced large increases in 5-HT thatexceeded its effects on dopamine. The microdialysis data forMDPV reported here are the first to show this drug fails toaffect extracellular concentrations of 5-HT in vivo at pharma-cologically relevant doses. The microdialysis data withmethylone agree with our previous findings that used lowerdoses of i.v. administered drug (i.e., 0.1 and 0.3 mg/kg i.v.) todemonstrate that methylone and mephedrone exhibit MDMA-like neurochemical effects in male rats (Baumann et al. 2012).In the present study, MDPV and methylone both increasedlocomotor activity. However, MDPV induced much larger in-creases in hyperactivity than methylone, even though thedrugs had similar effects on extracellular dopamine. Our

previous microdialysis experiments have shown that eleva-tions in extracellular dopamine in the nucleus accumbens aretightly correlated with the extent of locomotor activity pro-duced by stimulant drugs (Baumann et al. 2008, 2011;Zolkowska et al. 2009). Thus, one interpretation of the currentmicrodialysis and locomotor data is that elevations in extra-cellular 5-HTcan reduce locomotor stimulant effects mediatedby extracellular dopamine. Future studies could address thishypothesis by examiningmicrodialysis and locomotor activityresponses for other types of stimulant drugs in rodent models.

In summary, a direct comparison between MDPV andmethylone demonstrated that MDPV supports robust self-administration in rats. Although methylone also supportedself-administration, this drug is not as potent or efficaciousas MDPV, maintaining lower rates of intake and not showinga clear dose-effect function. Based on the findings presentedhere, it is tempting to speculate that elevations in extracellular5-HT produced by methylone can reduce its reinforcing ef-fects in rats, but this view may be an oversimplification. Therecent article by Creehan et al. (2015) is noteworthy because itcompared the self-administration of MDMA, methylone andmephedrone in female rats. These three Bempathogen^ drugsare non-selective transporter substrates that increase the re-lease of dopamine, norepinephrine and 5-HT in vitro(Baumann et al. 2012; Eshleman et al. 2013; Simmler et al.2013). Importantly, the 5-HT-releasing ability of these drugs ismore prevalent than their effects on dopamine in vivo(Baumann et al. 2008, 2012; Kehr et al. 2011; Wright et al.2012). Despite the shared mechanism of action among thesedrugs, Creehan et al. (2015) showed that mephedrone displayssignificantly greater reinforcing effects than either methyloneor MDMA. Such findings indicate that other factors in addi-tion to 5-HT release, including drug pharmacokinetics, effectson noradrenergic systems, or non-transporter sites of action,must influence the self-administration behavior of MDMA-like drugs. Finally, our self-administration data from rats areconsistent with the known misuse of MDPVand methylone inhumans (e.g., Johnson and Johnson 2014) and suggest thatMDPV is more apt to be abused in a compulsive mannerdue to its potent and selective dopaminergic effects.

Acknowledgments This research was supported by the Intramural Re-search Program of NIH, NIDA. The authors have no conflicts of interestto report. We dedicate this paper to our colleague and friend Steven R.Goldberg who passed away suddenly on November 25, 2014.

References

Aarde SM,Huang PK, CreehanKM,Dickerson TJ, TaffeMA (2013) Thenovel recreational drug 3,4-methylenedioxypyrovalerone (MDPV)is a potent psychomotor stimulant: self-administration and locomo-tor activity in rats. Neuropharmacology 71:130–40

Psychopharmacology

Aarde SM, Huang PK, Dickerson TJ, Taffe MA (2015) Binge-like acqui-sition of 3,4-methylenedioxypyrovalerone (MDPV) self-administration and wheel activity in rats. Psychopharmacology(Berl) 232:1867–77

Bauer CT, Banks ML, Blough BE, Negus SS (2013) Use of intracranialself-stimulation to evaluate abuse-related and abuse-limiting effectsof monoamine releasers in rats. Br J Pharmacol 168:850–62

Baumann MH (2014) Awash in a sea of ‘bath salts’: implications forbiomedical research and public health. Addiction 109:1577–9

Baumann MH, Clark RD, Rothman RB (2008) Locomotor stimulationproduced by 3,4-methylenedioxymethamphetamine (MDMA) iscorrelated with dialysate levels of serotonin and dopamine in ratbrain. Pharmacol Biochem Behav 90:208–17

BaumannMH, Clark RD,WoolvertonWL,Wee S, Blough BE, RothmanRB (2011) In vivo effects of amphetamine analogs reveal evidencefor serotonergic inhibition of mesolimbic dopamine transmission inthe rat. J Pharmacol Exp Ther 337:218–25

Baumann MH, Ayestas MA Jr, Partilla JS, Sink JR, Shulgin AT, DaleyPF, Brandt SD, Rothman RB, Ruoho AE, Cozzi NV (2012) Thedesigner methcathinone analogs, mephedrone and methylone, aresubstrates for monoamine transporters in brain tissue.Neuropsychopharmacology 37:1192–203

Baumann MH, Partilla JS, Lehner KR, Thorndike EB, Hoffman AF, HolyM, Rothman RB, Goldberg SR, Lupica CR, Sitte HH, Brandt SD,Tella SR, Cozzi NV, Schindler CW (2013) Powerful cocaine-like ac-tions of 3,4-methylenedioxypyrovalerone (MDPV), a principal constit-uent of psychoactive ‘bath salts’ products. Neuropsychopharmacology38:552–62

Bonano JS, Glennon RA, De Felice LJ, Banks ML, Negus SS (2014)Abuse-related and abuse-limiting effects of methcathinone and thes y n t h e t i c Bb a t h s a l t s ^ c a t h i n o n e a n a l o g smethylenedioxypyrovalerone (MDPV), methylone and mephedroneon intracranial self-stimulation in rats. Psychopharmacology (Berl)231:199–207

Bradbury S, Bird J, Colussi-Mas J, Mueller M, Ricaurte G, Schenk S(2014) Acquisition of MDMA self-administration: pharmacokineticfactors and MDMA-induced serotonin release. Addict Biol 19:874–84

Cameron KN, Kolanos R, Solis E Jr, Glennon RA, De Felice LJ (2013)Bath salts components mephedrone and methylenedioxypyrovalerone(MDPV) act synergistically at the human dopamine transporter. Br JPharmacol 168:1750–7

Creehan KM, Vandewater SA, Taffe MA (2015) Intravenous self-administration of mephedrone, methylone and MDMA in femalerats. Neuropharmacology 92:90–7

De La Garza R 2nd, Fabrizio KR, Gupta A (2007) Relevance of rodentmodels of intravenous MDMA self-administration to humanMDMA consumption patterns. Psychopharmacology (Berl) 189:425–34

Drug Enforcement Administration (2011) Schedules of controlled sub-stances: temporary placement of three synthetic cathinones in sched-ule I. Final order. Fed Regist 76:65371–65375

Drug Enforcement Administration (2013) Establishment of drug codesfor 26 substances. Final rule. Fed Regist 78:664–666

Drug Enforcement Administration (2014) National Forensic LaboratoryInformation System Special Report: Synthetic Cannabinoids andSynthetic Cathinones Reported in NFLIS, 2010–2013. U.S. DrugEnforcement Administration, Springfield, VA URL: https://www.nflis.deadiversion.usdoj.gov/DesktopModules/ReportDownloads/Reports/NFLIS_SR_CathCan_508.pdf

Eshleman AJ, Wolfrum KM, Hatfield MG, Johnson RA, Murphy KV,Janowsky A (2013) Substituted methcathinones differ in transporterand receptor interactions. Biochem Pharmacol 85:1803–15

Fantegrossi WE, Gannon BM, Zimmerman SM, Rice KC (2013) In vivoe f f e c t s o f a b u s e d ‘ b a t h s a l t ’ c o n s t i t u e n t 3 , 4 -methylenedioxypyrovalerone (MDPV) in mice: drug

discrimination, thermoregulation, and locomotor activity.Neuropsychopharmacology 38:563–73

Gatch MB, Taylor CM, Forster MJ (2013) Locomotor stimulant anddiscriminative stimulus effects of ‘bath salt’ cathinones. BehavPharmacol 24:437–47

German CL, Fleckenstein AE, Hanson GR (2014) Bath salts and synthet-ic cathinones: an emerging designer drug phenomenon. Life Sci 97:2–8

Johnson PS, Johnson MW (2014) Investigation of Bbath salts^ use pat-terns within an online sample of users in the United States. JPsychoactive Drugs 46:369–78

Karlsson L, Andersson M, Kronstrand R, Kugelberg FC (2014)Mephedrone, methylone and 3,4-methylenedioxypyrovalerone(MDPV) induce conditioned place preference in mice. Basic ClinPharmacol Toxicol 115:411–416

Kehr J, Ichinose F, Yoshitake S, Goiny M, Sievertsson T, Nyberg F,Yoshitake T (2011) Mephedrone, compared with MDMA (ecstasy)and amphetamine, rapidly increases both dopamine and 5-HT levelsin nucleus accumbens of awake rats. Br J Pharmacol 164:1949–58

Kesha K, Boggs CL, Ripple MG, Allan CH, Levine B, Jufer-Phipps R,Doyon S, Chi P, Fowler DR (2013) Methylenedioxypyrovalerone(Bbath salts^), related death: case report and review of the literature.J Forensic Sci 58:1654–9

Kiyatkin EA, Kim AH, Wakabayashi KT, Baumann MH, Shaham Y(2015) Effects of social interaction and warm ambient temperatureon brain hyperthermia induced by the designer drugs methylone andMDPV. Neuropsychopharmacology 40:436–45

Lopez-Arnau R, Martinez-Clemente J, Pubill D, Escubedo E, Camarasa J(2012) Comparative neuropharmacology of three psychostimulantcathinone derivatives: butylone, mephedrone and methylone. Br JPharmacol 167:407–20

Lopez-Arnau R, Martinez-Clemente J, Abad S, Pubill D, Camarasa J,Escubedo E (2014a) Repeated doses of methylone, a new drug ofabuse, induce changes in serotonin and dopamine systems in themouse. Psychopharmacology (Berl) 231:3119–29

Lopez-Arnau R, Martinez-Clemente J, Pubill D, Escubedo E, Camarasa J(2014b) Serotonergic impairment andmemory deficits in adolescentrats after binge exposure of methylone. J Psychopharmacol 28:1053–63

Marusich JA, Grant KR, Blough BE,Wiley JL (2012) Effects of syntheticcathinones contained in Bbath salts^ on motor behavior and a func-tional observational battery in mice. Neurotoxicology 33:1305–13

Panlilio LV, Katz JL, Pickens RW, Schindler CW (2003) Variability ofdrug self-administration in rats. Psychopharmacology (Berl) 167:9–19

Schenk S, Hely L, Lake B, Daniela E, Gittings D, Mash DC (2007)MDMA self-administration in rats: acquisition, progressive ratioresponding and serotonin transporter binding. Eur J Neurosci 26:3229–36

Schindler CW, Cogan ES, Thorndike EB, Panlilio LV (2011) Rapid de-livery of cocaine facilitates acquisition of self-administration in rats:an effect masked by paired stimuli. Pharmacol Biochem Behav 99:301–6

Seely KA, Patton AL, Moran CL, Womack ML, Prather PL, FantegrossiWE, Radominska-Pandya A, Endres GW, Channell KB, Smith NH,McCain KR, James LP, Moran JH (2013) Forensic investigation ofK2, Spice, and Bbath salt^ commercial preparations: a three-yearstudy of new designer drug products containing synthetic cannabi-noid, stimulant, and hallucinogenic compounds. Forensic Sci Int233:416–22

Simmler LD, Buser TA, Donzelli M, Schramm Y, Dieu LH, Huwyler J,Chaboz S, Hoener MC, Liechti ME (2013) Pharmacological charac-terization of designer cathinones in vitro. Br J Pharmacol 168:458–70

Watterson LR, Hood L, Sewalia K, Tomek SE, Yahn S, Johnson CT,Wegner S, Blough BE, Marusich JA, Olive MF (2012) The

Psychopharmacology

reinforcing and rewarding effects of methylone, a syntheticcathinone commonly found in Bbath salts^. J Addict Res Ther S9:002

Watterson LR, Kufahl PR, Nemirovsky NE, Sewalia K, Grabenauer M,Thomas BF, Marusich JA, Wegner S, Olive MF (2014) Potent re-warding and reinforcing effects of the synthetic cathinone 3,4-methylenedioxypyrovalerone (MDPV). Addict Biol 19:165–74

Wee S, Woolverton WL (2006) Self-administration of mixtures of fen-fluramine and amphetamine by rhesus monkeys. PharmacolBiochem Behav 84:337–43

World Health Organization (2015). WHO expert committee on drug de-pendence: thirty-sixth report. WHO Technical Report Series no.991. Geneva, Switzerland . URL: http://apps.who.int/iris/bitstream/10665/153834/1/WHO_TRS_991_eng.pdf?ua=1

Wright MJ Jr, Angrish D, Aarde SM, Barlow DJ, Buczynski MW,Creehan KM, Vandewater SA, Parsons LH, Houseknecht KL,Dickerson TJ, Taffe MA (2012) Effect of ambient temperature onthe thermoregulatory and locomotor stimulant effects of 4-methylmethcathinone in Wistar and Sprague-Dawley rats. PLoSOne 7:e44652

Wright TH, Cline-Parhamovich K, Lajoie D, Parsons L, Dunn M,Ferslew KE (2013) Deaths involving methylenedioxypyrovalerone(MDPV) in Upper East Tennessee. J Forensic Sci 58:1558–62

Zolkowska D, Jain R, Rothman RB, Partilla JS, Roth BL, Setola V,Prisinzano TE, Baumann MH (2009) Evidence for the involvementof dopamine transporters in behavioral stimulant effects ofmodafinil. J Pharmacol Exp Ther 329:738–46

Psychopharmacology